The proposed research has as its overall goal to define mechanisms that underlie normal retinal attachment, with the aim that better methods of restoring the interaction between the neural retina and its supporting retinal pigment epithelium (RPE) may be developed, and thus loss of visual function resulting from retinal detachment ameliorated. Little is known about how the structural components of the RPE-neural retina interface are generated, in terms of specific signaling events and their sequence of activation in development. The proposed experiments focus specifically on the role of the epithelium, and will make use of a strain of mice that is deficient in the cell cycle regulatory protein p27(Kip1). Aside from retinal proliferative defects and limited foci of retinal dysplasia, these animals exhibit broad areas of retinal detachment, evident as a loss of interdigitation between photoreceptor outer segments and RPE microvilli. Aim 1 will test the hypothesis that p27(Kip1) plays a role in establishment and maintenance of the RPE-neural retina interface that is independent of its involvement in epithelial cell cycle regulation. For these experiments, two strains of knock-in mice will be examined using established light and electron microscopic methods. The first strain (p27CK-) contains point mutations in the p27(Kip1) binding sites for cyclins and cyclin-dependent kinase (CDK), resulting in a protein that is unable to act as a cell cycle inhibitor, but nevertheless retains its cytoskeleton regulatory properties. In the second strain (p27S10A), the major phosphorylation site regulating p27(Kip1) protein export to the cytoplasm has been altered, yielding a protein that can carry out its function as a negative regulator of cell proliferation within the nucleus but is unable to influence the actin cytoskeleton. The possibility that the p27(S10A) mutation causes the induction of retinal detachment, and that p27CK- rescues detachment resulting from p27(Kip1) gene ablation will be examined. Finally, the possibility that retinal detachment is due to increased RPE cellular or nuclear density per se will be examined by crossing p27(Kip1)-null mice with mice lacking the positive cell cycle regulatory gene cyclin D1. Aim 2 will examine whether the retinal detachment we observe in p27(Kip1)-null mice results, at least in part, from a loss of protein function specifically within the epithelium. To test this, the Cre-lox system for conditional gene targeting will be implemented to ablate the p27(Kip1) gene selectively in RPE cells. [unreadable] [unreadable] [unreadable]